Antibody-drug conjugates present challenges around compound registration and property prediction.  Antibodies are registered as amino acid sequences. Calculation of properties like molecular weight is straightforward. Small molecules also have a variety of formats for registration of compounds and software tools for property calculation. When small molecules and antibodies are conjugated, the problems of registration and property calculation becomes more complex. We will discuss approaches to solve these problems.

Immunoassay platforms are essential tools in bioanalytical studies, recognized for their high sensitivity and specificity, minimal sample volume requirements, and compatibility with 384-well plates. However, the complexity involved in assay development is a challenge in daily operations and the extensive protocols often hinder automated sample analysis. To overcome these challenges and enhance integration of immunoassays into our bioanalytical workflows, we are adopting new digitalization and integrated automation strategies for both assay development and sample analysis. This presentation will outline our current end-to-end platforms, encompassing both in-silico and wetlab aspects, as well as our future initiatives in digitalization, AI, and automation.

I will present an envisioned automated analytical workflow for RPIP-HPLC chromatograms, covering baseline inference, retention-time alignment, peak detection, deconvolution, and quantification. Parameter settings are optimized via Bayesian search. Every step—from raw-file ingestion and versioned configurations to QC metrics, anomaly flags, a cautious LLM-assisted reviewer-triage step to aid manual review and decision-making—is logged with immutable provenance, enabling auditability and supporting GxP compliance. We target cross-modal applicability without prescribing instrument-specific workflows.

Microfluidic electrophoresis is a powerful characterization technique for both novel protein-based therapeutics and protein biomarkers. We will discuss innovative methods that use both size-based and charge-based automated microfluidic electrophoresis to analyze different types of proteins and how this translates to the drug discovery and development process. We will also discuss how the results from these findings can be paired with artificial intelligence and how our predictive analysis method can be used for reagent manufacturing protocols for microfluidics applications.

We present an integrated high-throughput workflow combining predictive modeling and automated purification to accelerate multispecific development. Leveraging data-driven parameter optimization and digital analytics, this scalable strategy enhances yield, robustness, and process consistency across diverse molecule classes including DuetMab and TITAN formats. Our standardized platform integrates walk-away automated affinity purification with multi-sample SEC, significantly reducing purification steps and in-process controls for large-scale manufacturing. This approach spans multiple molecule classes (IgG1, mIgG2a, VHH, Fab, VHH-Fc) and volume ranges (30–1000 mL), demonstrating significant efficiency gains and enabling rapid transition from design to manufacturing while maintaining quality standards for complex biotherapeutics.

Monoclonal antibody product quality is influenced by a complex interplay of the manufacturing process, formulation, and inherent structural attributes of the molecule. To better understand these relationships, an in silico data lake was created using state-of-the-art structure and property prediction tools, which enabled development of an unsupervised ML model of the biophysical feature space. This pipeline offers critical insight into how biophysical properties contribute to various colloidal stability-related outcomes in CMC development.

Explore how digital tools and data automation are advancing the monitoring, evaluation, and enhancement of analytical methods for separation, impurity, and potency. Innovative strategies for integrating data and harnessing real-time insights will be showcased, enabling streamlined workflows and driving continuous improvement across the entire analytical lifecycle—elevating data quality, operational efficiency, and method performance in biotherapeutic analytics.

The escalating demand for patient-friendly subcutaneous administration necessitates the development of high-concentration liquid biologic formulations. Yet, predicting their complex protein behaviors, including viscosity and aggregation, presents significant developability challenges. To address this, we have developed an integrated and automated early screening and selection workflow. This robust process leverages high-throughput, low-mass assays in conjunction with powerful in silico developability assessments. By proactively evaluating critical solution parameters and predicting potential risks across diverse molecule formats, our platform empowers researchers to make informed decisions and optimize the manufacturability and stability of biologics earlier in the drug discovery pipeline.

Modern biotherapeutic pipelines demand truly high-throughput, automated developability assessment to evaluate increasing candidate volumes efficiently. This presentation explores implementing standardized, automated assay platforms to generate comprehensive, high-quality developability datasets. By establishing this high-throughput developability data collection, we enable early identification of developability risks before costly downstream manufacturing issues arise, and generate the datasets essential for training robust machine-learning models.

This talk introduces a data-centric platform that combines robotics, high-throughput biophysical characterization, and decision-ready visualization. Automated liquid handling standardizes sample prep and scales throughput, while multiplexed assays quantify biophysical and stability parameters to identify risks early. Case studies show shorter cycles, improved data quality, and stronger portfolio decisions, with sample traceability, method validation, and seamless integration across automation platforms and informatics systems.

We will share strategies leveraged in our group to enable forward compatibility of our platforms. We will address automation, data capture and custom engineering solutions to adopt our instrumentation to our science.

Advances in automation and AI have revolutionized the field of biologics discovery. Quantity of data is exponentially increasing, and ML architectures are rapidly improving. Key to leveraging this technological revolution lies in accessibility of data and AI. I will talk about our efforts at Lilly in developing an integrated digital platform that allows us to fully leverage experimental and data science toward improved decision-making and accelerated DMTA cycles.